agentleFS
Sign inStart free
yiyaw-labyiyaw-lab/agent-armor/prompt/SKILL.md

prompt

Skill1 starsChanged 5 months agoMITView on GitHub

Personal prompt library + prompt compiler. Save frequently used prompts as parameterized templates, invoke them by name, or refine a draft prompt into a more effective, token-lean version — informed by TASTE.md so the refined prompt asks for what the user will actually accept, not just what they typed. Modes: <name> | save | list | refine.

20 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use brainstorm, init and 1 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • nicolasestremSkill

    brainstorm

    Structured idea generation and solution exploration with systematic analysis.

    Also in 4 projects that use this

  • MicrosoftSkill

    init

    Also in 3 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 18 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write the file, report: name, when-to-use, and an invocation example.

    line 14
  2. Very specific:

    Acceptance criteria from TASTE.md: Read ~/.claude/TASTE.md ACTIVE rules and bake the relevant ones into the prompt explicitly — e.g.

    line 22
  3. Very specific:

    Never execute a refine result without the user sending it.

    line 31

“Common” is measured: kept by at least 6 in 10 of the 18 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

The library lives in ~/.claude/prompts/, one file per prompt: frontmatter (name, when — one line on when to reach for it) + body = the template, with {placeholders} for the parts that vary.

/prompt <name> [args]

Read ~/.claude/prompts/<name>.md, fill {placeholders} from args and session context, then EXECUTE the template as if the user had typed it in full. Infer missing placeholders from context; ask only if genuinely ambiguous. (This is the token saver: the user types 3 words, the agent runs the 80-word version that actually works.)

/prompt save [name] [text]

  • Source: the given text, or the user's most recent substantive prompt this session.
  • Distill to a reusable template: replace session-specific details with {placeholders}; KEEP the phrasing that made it effective (constraints, the bar, the verification ask).
  • Write the file, report: name, when-to-use, and an invocation example.

/prompt list

Table: name | when to use | example invocation. If the library is empty, say so and point at save/refine.

/prompt refine <draft | "last">

Compile the prompt for effect-per-token. Diagnose, then rebuild:

  1. Goal, not artifact. State the outcome wanted, not just the deliverable. (The user's revealed pattern: the real bar arrives as round-2 corrections — front-load it instead.)
  2. Acceptance criteria from TASTE.md. Read ~/.claude/TASTE.md ACTIVE rules and bake the relevant ones into the prompt explicitly — e.g. modern design idiom, verified-not-claimed side effects, system-level scope, loose ends closed. This is the difference between what the user types and what they'll accept.
  3. Cut filler. Pleasantries, hedges, context the agent already has, double-asks.
  4. Structure. Constraints and output format as terse bullets; one primary ask per prompt.
  • Output: BEFORE (word count) → AFTER (word count), then a 2–3 bullet "what changed and why."
  • Rule: the refined prompt must not be longer than the original unless missing acceptance criteria justify the growth — effectiveness first, tokens second. Never pad.
  • End by offering /prompt save <suggested-name> if the shape looks reusable.

Rules

  • Templates are the user's voice, not corporate prompt-speak — terse, direct, no "please act as."
  • Never execute a refine result without the user sending it; refine produces text, not action.
  • The library is local. Publishing it is a disclosure decision (prompts can leak project strategy).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free